The Official Kaveri Review Thread (A10-7850K, etc)

[Enigmoid]

Core i7 4770K has the same TDP value of 84W as Core i5 4430. Core i7 4770K has higher Power Consumption than Core i5 4430. With the same thinking, Intel is cheating in the TDP value of Core i7. Do you get it now ???
Shivansps compared the TOTAL Platform power consumption of different 45W TDP SKUs in Reviews to come to the conclusion that AMD is cheating with Kaveri 45W TDP value simple because another 45W TDP SKU has lower power consumption than Kaveri.
As we have said so many times here before, TDP is not Power Consumption. Two different 45W TDP SKUs can have different power consumptions. Same applies to Intel SKUs both in Desktop, Mobile and Server.

Well then stop looking at tdp and performance and comparing to power consumption on mobile. Performance/tdp is worthless. You agree. Then calculated performance on mobile when that data is extrapolated from performance/tdp is also useless.

The end result is that kaveri (at a certain tdp) performs much better than richland but consumes more power.

There was throttling with some P4 models. Did your hear about that?

I don't see how this is relevant. Turbo should be fully attainable.

1. The A8 7600 has a configurable TDP - meaning user can choose between 45W and 65W. Is it your opinion that the CPU and GPU clocks should never drop bellow maximum values no matter what TDP you choose?

2. If "boost" is to be "fully and completely available", what is the point of calling it boost? Wouldn't it be nominal speed?

1. A8-7600 has different clocks for each TDP.

2. Boost makes perfect sense for situations where there are thermal limitations (mobile) or deactivating several cores to attain a higher frequency on the remaining.

3. Mobile is different from desktop (which I specifically said), mainly due to cooling requirements (more stringent tdp requirements).

4. That is the reason why real world tests are necessary. With boost and dropping clocks you cannot look at a prime and furmark test (with dropped clocks) and then at performance (at boost clocks) to get any measure of efficiency and extrapolate to mobile.

Toms and techreport did that and showed that there was not a huge change in efficiency.

My problem is not with the tdp business but with the extrapolation to mobile based on tdp.

Edit: I have no problem with limiting tdp (max power consumption) to fit SFF. But that is greatly difference from a change in efficiency. ex) The 4770k could be limited to 70W. This would not affect anything but extreme cases of power draw (and drop clocks). However, this would NOT change the efficiency of the CPU, nor would it affect how the CPU performs in mobile.

 

[coercitiv]

1. A8-7600 has different clocks for each TDP.

Yes it does, and the further down you go in TDP, the less likely you are to be able to maintain boost clocks in full load (both CPU and iGPU).

2. Boost makes perfect sense for situations where there are thermal limitations (mobile) or deactivating several cores to attain a higher frequency on the remaining.

No it doesn't, you've just been had by the marketing dep. Boost is the natural step you would take after being able to check&control how much power the chip uses at a certain time. It's a natural move and makes sense in any scenario where performance is needed.

3. Mobile is different from desktop (which I specifically said), mainly due to cooling requirements (more stringent tdp requirements).

When we are talking about a 45W TDP CPU the border between desktop and mobile is hardly there anymore. My mobile CPU has a 47W TDP.

Edit: I have no problem with limiting tdp (max power consumption) to fit SFF. But that is greatly difference from a change in efficiency. ex) The 4770k could be limited to 70W. This would not affect anything but extreme cases of power draw (and drop clocks). However, this would NOT change the efficiency of the CPU, nor would it affect how the CPU performs in mobile.

Limiting TDP leads to dropping clocks in certain scenarios, and changing clocks changes efficiency. It also affects how the CPU performs in mobile.

 

[Phynaz]

There was throttling with some P4 models. Did your hear about that?

How is going back a decade relevant?

 

[AtenRa]

emmm....yes and no. TDP(heat) = max* (electrical)power consumption for real world loads. IOW if one sees APU/CPU electrical power consumption exceed TDP value , then one might start screaming for fraud.

* As to (some)intel's CPUs - those have some transient peak power "feature" thus they might exceed the TDP value for short (state changing etc) bursts, if there is enough temperature headroom.


edited

Again for the BILLIONTH time, TDP is not the max power consumption of the CPU.

http://www.intel.com/content/dam/doc/white-paper/resources-xeon-measuring-processor-power-paper.pdf

TDP (Thermal Design Power)
Intel defines TDP as follows:
The upper point of the thermal profile consists of the Thermal Design
Power (TDP) and the associated Tcase value. Thermal Design Power (TDP) should be used for processor thermal solution design targets. TDP is not the maximum power that the processor can dissipate. TDP is measured at maximum TCASE.
1.The thermal profile must be adhered to to ensure Intel’s reliability requirements are met.

Also, how AMD defines TDP.

TDP . Thermal Design Power. The thermal design power is the maximum power a processor can draw for a thermally significant period while running commercially useful software. The constraining conditions for TDP are specified in the notes in the thermal and power tables.”
Notes: - TDP is measured under the conditions of all cores operating at CPU COF, Tcase Max, and VDD at the voltage requested by the processor. TDP includes all power dissipated on-die from VDD,VDDNB, VDDIO, VLDT, VTT and VDDA.
- The processor thermal solution should be designed to accommodate thermal design power (TDP) at Tcase,max.TDP is not the maximum power of the processor.

The only "fraud" here is that people consistently doesn't understand that TDP is not max power consumption of the CPU, both AMD and Intel specifically outline that TDP is not the max power consumption.

 

[9enesis]

Again for the BILLIONTH time, TDP is not the max power consumption of the CPU.

(facepalm) so....answer then: why even put TDP rating on CPUs if power consumption may exceed it ?

did you even read what you post?:biggrin:

what do you think this means? :

[QUOTE=AtenRa;36000621] TDP is not the maximum power that the processor can dissipate

TDP . Thermal Design Power. The thermal design power is the maximum power a processor can draw for a thermally significant period while running commercially useful software[/QUOTE]

 

[AtenRa]

Well then stop looking at tdp and performance and comparing to power consumption on mobile. Performance/tdp is worthless. You agree. Then calculated performance on mobile when that data is extrapolated from performance/tdp is also useless.

The end result is that kaveri (at a certain tdp) performs much better than richland but consumes more power.

TDP is very relevant especially in Mobile. What we see with Kaveri is that at the same TDP and that means at the same THERMAL restrictions that applies in Mobile devices, it can have consistently higher performance than Richland. That means it can have higher base clocks and higher Turbo clocks. It also means that it will also be able to keep Turbo clocks for longer periods. All that translates in to extremely higher single thread performance than Richland, higher multithread performance and higher iGPU performance.

So to sum up,
I used the lower TDP 45W Desktop SKU to show that a 35W TDP Mobile Kaveri will have a substantial performance gain over the same TDP Richland because TDP matters the most in Mobile. I have never talked about power consumption but performance at the same TDP level.

 

[Shivansps]

Im not sure why its so hard to understand, a 45W TDP cpu can set the turbos to work over the 45W TDP as long the cooler supports it, its not really a 45W TDP cpu if it can maintain turbos all the way up to 85W because of a good cooler, im not talking about power here, Intel mobile and ULV cpus do that, configurable TDP does not means much unless it also have an effect on the hidden TDP, say 45W and up to 65W, them 65W and up to 85W.

All of that means that a 35W Kaveri Mobile will be unable to do the same, it will be constrained as intel CPU are, especially with OEMs looking to go cheaper on everything.

thats because i do not belive that the A8-7600 is getting that performance in just 45W TDP window.

 

[9enesis]

Im not sure why its so hard to understand, a 45W TDP cpu can set the turbos to work over the 45W TDP as long the cooler supports it.

because it is NOt so, ok? look , when you design a chip or something , you have to provide specification for potential customers , one of the parametres important is peak power draw - that would be TDP . period. why TDP = peak or continuous power draw? BECAUSE - there is such a thing as the second law of thermodynamics - which in this case states : that electrical energy you put in - all comes out as heat - or....the amont of power on input = heat in Watts disipated. it is not something i invented - this is how nature works!


P.S. i'm sorry , english is not my native. i hope you get the idea.

 

[Chiropteran]

that electrical energy you put in - all comes out as heat - or....the amont of power on input = heat in Watts disipated. it is not something i invented - this is how nature works!

That is a simplistic and flawed view.

Put a pot of cold water on a stove burner. Turn the burner to maximum for 20 seconds. Notice that the water still is just water, and not boiling. Now reduce the burner to about 50%, and wait. Eventually the water will boil.

Note that while 50% output sustained over time was sufficient to boil the water, a burst of 100% for a short period was not.

This is how nature works, and this is why a CPU may use more power than the listed TDP would imply over a limited period of time.

 

[Erenhardt]

OMG. Now we have two threads about TDP and power consumption. (see what I did there? )
Could you argue that in dedicated thread?

 

[9enesis]

That is a simplistic and flawed view.

Put a pot of cold water on a stove burner. Turn the burner to maximum for 20 seconds. Notice that the water still is just water, and not boiling. Now reduce the burner to about 50%, and wait. Eventually the water will boil.

Note that while 50% output sustained over time was sufficient to boil the water, a burst of 100% for a short period was not.

This is how nature works, and this is why a CPU may use more power than the listed TDP would imply over a limited period of time.

LOL , so what , that it did not boil at that few second 100% burst ? you still DID exceeded the TDP rating , right? :biggrin:

it cannot use more power because of specified TDP rating, otherwise you might end up with dead MoBo(can not deliver the power CPU consumes) or burned CPU because of not enough cooling later.

i really do not get why it is so hard to understand basic things...

 

[AtenRa]

LOL , so what , that it did not boil at that few second 100% burst ? you still DID exceeded the TDP rating , right? :biggrin:

No it did not exceed the TDP rating because the water didnt boil. Thus it kept the processor thermals within Tcase Max temperature.

 

[9enesis]

No it did not exceed the TDP rating because the water didnt boil. Thus it kept the processor thermals within Tcase Max temperature.

(facepalm) i suggest you to go and read a physics text book, ok?

yes he DID - just because he did not reach the max temperature - doesn't mean anything.somehow you both assuming that CPU is always cool at the start up, why is that?

the CPU or APU is effectivly a heater , that means that all electrical power is converted into heat. Do you understand what that means?

put it this way: assume that your stock CPU is rated at some 65W and you cool it by liquid nitrogen - IOW you put in 65W electrical power and it never gets hot....now what?

 

[AtenRa]

(facepalm) i suggest you to go and read a physics text book, ok?

yes he DID - just because he did not reach the max temperature - doesn't mean anything.

the CPU or APU is effectivly a heater , that means that all electrical power is converted into heat. Do you understand what that means?

Your problem is that you take TDP as power consumption.

He used 150W for 5 secs and the water didnt boiled (The water didnt reach Tcase Max) thus he didnt breached the TDP rating of 65W.

In the second time he used 75W for 2 minutes and the water started to boil thus it reach Tcase max temperature and breached the TDP limit of 65W TDP.

So you see that TDP is not max power consumption.

 

[9enesis]

Your problem is that you take TDP as power consumption.

He used 150W for 5 secs and the water didnt boiled (The water didnt reach Tcase Max) thus he didnt breached the TDP rating of 65W.

In the second time he used 75W for 2 minutes and the water started to boil thus it reach Tcase max temperature and breached the TDP limit of 65W TDP.

So you see that TDP is not max power consumption.

oh FFS ! what has this to do with temperature ?! you drop out that period while initial heating , start with water already boiling -will it make easyer for you to understand then?

also you do not seem to understand that in order to exceed the TDP rating your MoBo and your PSU have to be able to deliver that extra amount of power !

Now go try to make such a TDP "trick" in your laptop....see how it will end!

 

[Erenhardt]

oh FFS ! what has this to do with temperature ?! you drop out that period while initial heating , start with water already boiling -will it make easyer for you to understand then?

also you do not seem to understand that in order to exceed the TDP rating your MoBo and your PSU have to be able to deliver that extra amount of power !

Now go try to make such a TDP "trick" in your laptop....see how it will end!

Thermal design =/= electrical design

In thermals you have thermal capacity which is the amount of energy that can be stored for further dissipation. Electrical power spike will occur, but will be not reflected in heat dissipation right away.
Best explanation I could provide with my broken English

 

[9enesis]

Thermal design =/= electrical design

it is(because CPU is just a heater by itself) - with an exception that TDP is "inert" and EDP is "instantaneous"...if i may call it so.once you heat it up to some temperature x - there is no more "heat storage" available.

Also - there is no point to make 100W EDP for a CPU with 50W TDP....it is all about the cost of peripherials , such as motherbords and PSUs, batteries.

In thermals you have thermal capacity which is the amount of energy that can be stored for further dissipation. Electrical power spike will occur, but will be not reflected in heat dissipation right away.
Best explanation I could provide with my broken English

yes there is some "hysteresis" for heat ,if we may call it so and Intel indeed exploits it - also they had clearly stated , that this feature is configurable and thay have also stated by how much would that increase the power draw - it is all in the specs.also it only works for short bursts when the CPU is cool enough.

the point is that TDP is manufacturer defined thing , thus manufacturer has to make sure their chip holds that.

 

[AtenRa]

oh FFS ! what has this to do with temperature ?!

Well, TDP by definition has everything to do with Thermals and Temperatures. The name itself is Thermal Design Power. It is the power the cooling solution should be able to dissipate in order your processor will never breach the Tcase max temperature.

 

[Phynaz]

That is a simplistic and flawed view.

Put a pot of cold water on a stove burner. Turn the burner to maximum for 20 seconds. Notice that the water still is just water, and not boiling. Now reduce the burner to about 50%, and wait. Eventually the water will boil.

Before you consider a view as simplistic and flawed consider this:
The heat capacity of a pot of water in relation to the heat source is immensely greater than that of a heatsink to a CPU. For your analogy to work it would have to be close to a thimble of water.

http://en.wikipedia.org/wiki/Heat_capacity

 

[9enesis]

Well, TDP by definition has everything to do with Thermals and Temperatures. The name itself is Thermal Design Power. It is the power the cooling solution should be able to dissipate in order your processor will never breach the Tcase max temperature.

LOL....no - wrong again! thermals(TDP) is about heat not about temperature.
yes, there is a difference in case you did not know that

 

[AtenRa]

The following graph clearly illustrates that A10-6700T 45W TDP Richland reaches Tcase Max after approximately 35 second and the processor reduces the power(gets slower) in order to keep it within the TDP rating of 45W and bellow Tcase Max temperature.

On the other hand, the Kaveri 45W TDP A8-7600 keeps the same power for the duration of the benchmark run. That indicates that the processor didnt reach the Tcase Max thermal limit and was able to use more power thus higher performance and higher power usage.

So people for the last time, TDP is not the max power usage. You never should compare the TDP rating with CPU power consumption.

 

[AtenRa]

LOL....no - wrong again! thermals(TDP) is about heat not about temperature.
yes, there is a difference in case you did not know that

The temperature is related to Tcase Max.

 

[9enesis]

The temperature is related to Tcase Max.

so what ? look , i give up. it is not like i am going to convice you anyway , am i ?:biggrin:

believe whatever you want to believe

 

[9enesis]

On the other hand, the Kaveri 45W TDP A8-7600 keeps the same power for the duration of the benchmark run. That indicates that the processor didnt reach the Tcase Max thermal limit and was able to use more power thus higher performance and higher power usage.

(facepalm)that is just bollocks. prove it first.

 

[Abwx]

Im not sure why its so hard to understand, a 45W TDP cpu can set the turbos to work over the 45W TDP as long the cooler supports it, its not really a 45W TDP cpu if it can maintain turbos all the way up to 85W because of a good cooler, im not talking about power here, Intel mobile and ULV cpus do that, configurable TDP does not means much unless it also have an effect on the hidden TDP, say 45W and up to 65W, them 65W and up to 85W.

All of that means that a 35W Kaveri Mobile will be unable to do the same, it will be constrained as intel CPU are, especially with OEMs looking to go cheaper on everything.

thats because i do not belive that the A8-7600 is getting that performance in just 45W TDP window.

Can you contradict what is written below..?..

On Hardware.fr test of the 7850K the APU comsumption
including VRMs losses was 55% of the plateform total
comsumption when benching with Fritzchess with 4 threads,
ratio must be much lower with a 45W TDP APU given that it
operate at a point where the PSU has even lower efficency,
likely slightly below 50% of the total comsumption will be
drained by such an APU.

On the TReport test the PSU has about 75% efficency at
80W input so only 60W goes to the plateform wich correlate
the 45W TDP of the APU.

Edit : Tomshardware test of the 7600 is made with
TDP configured at 65W............